Network loader

[ramosv]

Added dataset from tcga-brca:

Initial data dimensions:

miRNA: (1078, 503)
RNA: (775, 20532)
Methylation:(783, 20106)
Clinical: (1097, 18)
PAM50: (1087,)

Feature Selection Methods
Performed separately on Methylation and RNA datasets (top 1,000 features each):

• Unsupervised:

  • Variance Filter (highest variance)
  • Autoencoder Weights (largest weights)

• Supervised:

  • ANOVA F-test
  • RandomForest Feature Importance

also set up network loader to load the networks generated by SmCCNet from these feature selection.
More details for the data preprocessing in the README inside tcga_brca

[Copilot]

Pull Request Overview

This PR integrates TCGA BRCA dataset updates with enhanced feature selection methods, adds a NetworkLoader utility, and updates documentation and versioning to reflect the changes.

• Added RandomForest-based feature selection support and updated module imports.
• Implemented phenotype preprocessing in SmCCNet and introduced a new NetworkLoader class for network file management.
• Updated README, notebook examples, and CHANGELOG for consistency.

Reviewed Changes

Copilot reviewed 33 out of 41 changed files in this pull request and generated 1 comment.

Show a summary per file

File	Description
bioneuralnet/utils/init.py	Updated preprocess import to include RandomForest selection
bioneuralnet/external_tools/smccnet.py	Added phenotype_df validation and logger initialization
bioneuralnet/datasets/tcga_brca/README.md	Expanded and clarified TCGA BRCA data preprocessing and feature selection details
bioneuralnet/datasets/network_loader.py	Introduced new utility class for loading bundled network files
bioneuralnet/datasets/dataset_loader.py	Modified data loading to support different feature selection methods and file naming conventions
bioneuralnet/datasets/init.py	Updated all to export NetworkLoader
bioneuralnet/init.py	Updated version and module exports
README.md	Minor version update display
Cancer_example.ipynb	Expanded example with full pipeline demo for TCGA BRCA
CHANGELOG.md	Revised changelog to reflect version update and release notes

Files not reviewed (8)

• MANIFEST.in: Language not supported
• bioneuralnet/datasets/networks/brca_smccnet_ae/size_13_net_2.csv: Language not supported
• bioneuralnet/datasets/networks/brca_smccnet_rf/size_14_net_2.csv: Language not supported
• bioneuralnet/datasets/networks/brca_smccnet_rf/size_14_net_4.csv: Language not supported
• bioneuralnet/datasets/networks/brca_smccnet_rf/size_21_net_3.csv: Language not supported
• bioneuralnet/datasets/networks/brca_smccnet_var/size_14_net_2.csv: Language not supported
• bioneuralnet/datasets/networks/brca_smccnet_var/size_14_net_3.csv: Language not supported
• bioneuralnet/datasets/tcga_brca/brca_pam50.csv: Language not supported

Comments suppressed due to low confidence (1)

bioneuralnet/datasets/dataset_loader.py:52

• [nitpick] The file naming convention in dataset_loader (e.g., 'brca_mirna.csv') differs from the uppercase style referenced in the README. Consider standardizing the naming convention for clarity.

self.data["brca_mirna"]   = pd.read_csv(folder / "brca_mirna.csv",   index_col=0)

[Copilot]

Typo 'realease' should be corrected to 'release'.

Suggested change

	- **New realease**: A new release will include documentation for the other updates. (1.1.0)
	- **New release**: A new release will include documentation for the other updates. (1.1.0)